Typically, a packet mobile communication network requires the unobstructed call origination from mobile stations, call termination to mobile stations, and continuation of packet communication that is in progress despite the movement of the mobile stations within the network. For this reason, information such as the positions of mobile stations and the closest radio base station is constantly managed and updated as necessary in a packet mobile communication network. When the number of mobile terminals that are managed in a portable telephone system is extremely large, the centralized management of information of each individual mobile station leads to an increase in the amount of management traffic that must be transferred inside the network, and in addition, results in considerable time expended for the switching of packet transmission paths that accompanies movement. For this reason, techniques are typically employed wherein packet transmission devices and positional management servers are normally arranged hierarchically, and the managed traffic for movement within each hierarchy is localized within the hierarchy.
In the case of a mobile communication network based on, for example, the specifications of the GPRS (General Packet Radio Service) of 3GPP (3rd Generation Partnership Project), the packet transmission device of the highest level is referred to as the GGSN (Gateway GPRS Support Node), the packet transmission device of the next level is referred to as the SGSN (Serving GPRS Support Node), the packet transmission device of the next level is referred to as the RNC (Radio Network Controller) or BSC (Base Station Controller), and the radio base stations are arranged at the lowest level. The packet communication scheme in the GPRS is described in detail in the technical specification 3GPP TS 23.060, “General Packet Radio Service (GPRS); Service Description; Stage 2.”
In GPRS, when mobile stations are normally connected to an outside service provider, packet communication is performed, the following procedures are taken:
1. The mobile station selects the radio base station for which communication conditions are best and establishes a data link.
2. The mobile station establishes a data link to the RNC/BSC to which the radio base station belongs.
3. The mobile station establishes a data link to the SGSN to which the RNC/BSC belongs.
4. The mobile station reports to the SGSN the name of the service provider to which it is requesting connection. This name is referred to as the APN (Access Point Name).
5. The SGSN selects the appropriate GGSN in accordance with the reported APN and the subscriber information of the mobile station, and establishes a data link between the SGSN and the GGSN.
6. The mobile station begins to communicate packets with the outside service provider by way of the radio base station, the RNC/BSC, the SGSN, and the GGSN that have been determined in this way.
7. When the mobile station moves during communication, communication is continued by appropriately reselecting the connecting radio base station, RNC/BSC, and SGSN.
8. When communication is completed, the mobile station releases each of the data links.
In this way, when a mobile station connects to a service provider outside the mobile communication network in GPRS, and performes packet communication, the packet transmission path within the mobile communication network is substantially optimized. This is because the node to an outside service provider is fixed in the GGSN that is at the apex of the hierarchy of packet transmission devices. A similar type of hierarchical architecture is also employed in packet mobile communication systems other than GPRS.
However, constraints that force the transmission path to pass by way of the apex of the hierarchy of packet transmission devices when the communication partner of a mobile station is within the same mobile communication network results in the problem of redundancy in the packet transmission path. For example, in a packet communication services such as voice communication or instant messaging, the bulk of traffic is between mobile stations that are at relatively close positions within the same mobile communication network. Even though the packet transmission path is optimized if packets are returned at packet transmission devices that are at a low level in the hierarchy at this time, when the packet transmission path is restricted such that the packets pass by way of the packet transmission device that belongs to the highest-ranked level, network resources are needlessly consumed. In some services that are used by the mobile station, moreover, failure to route packets by way of a specific packet transmission devices may cause problems relating to levying charges i.e., optimization of packet transmission paths cannot always be achieved.